Ryan Liu

Email: ryanliu@princeton.edu   LinkedIn: ryanchenliu  GitHub: theryanl  Google Scholar: link   CV: pdf

Hi! I am a rising fourth year PhD student at Princeton Computer Science advised by Tom Griffiths. The central focus of my research is how large language models can transform how our society communicates and learns information. Previously, I was a Masters student at Carnegie Mellon working with Nihar Shah on solving central problems in conference peer review.

I am happy to chat about my current research and future opportunities! Please contact me via email at ryanliu@princeton.edu.

Papers

• Ads in AI Chatbots? An Analysis of How Large Language Models Navigate Conflicts of Interest
  Addison J. Wu*, Ryan Liu*, Shuyue Stella Li, Yulia Tsvetkov, and Thomas L. Griffiths
  Preprint 2026 [arXiv]


• Large Language Models Develop Novel Social Biases Through Adaptive Exploration
  Addison J. Wu*, Ryan Liu*, Xuechunzi Bai, and Thomas L. Griffiths
  ICML 2026 Oral [arXiv]


• Cognitive Models and AI Algorithms Provide Templates for Designing Language Agents
  Ryan Liu*, Dilip Arumugam, Cedegao E Zhang, Sean Escola, Xaq Pitkow, and Thomas L. Griffiths
  Preprint 2026 [arXiv]


• Levels of Analysis for Large Language Models
  Alexander Ku, Declan Campbell, Xuechunzi Bai, Jiayi Geng, Ryan Liu, Raja Marjieh, R. Thomas McCoy, Andrew Nam, Ilia Sucholutsky, Veniamin Veselovsky, Liyi Zhang, Jian-Qiao Zhu, and Thomas L. Griffiths
  Philosophical Transactions of the Royal Society A 2026 [arXiv]


• Evaluating Language Models’ Evaluations of Games
  Katherine M. Collins, Cedegao E. Zhang, Graham Todd, Lance Ying, Mauricio Barba da Costa, Ryan Liu, Prafull Sharma, Adrian Weller, Ionatan Kuperwajs, Lionel Wong, Joshua B. Tenenbaum, and Thomas L. Griffiths
  ICLR 2026 [arXiv]


• RLHS: Mitigating Misalignment in RLHF with Hindsight Simulation
  Kaiqu Liang, Haimin Hu, Ryan Liu, Thomas L. Griffiths, and Jaime Fernández Fisac
  ACL 2026 Findings [arXiv]


• Are Large Language Models Sensitive to the Motives Behind Communication?
  Addison J. Wu*, Ryan Liu*, Kerem Oktar*, Theodore R. Sumers, and Thomas L. Griffiths
  NeurIPS 2025 [arXiv]


• Accumulating Context Changes the Beliefs of Language Models
  Jiayi Geng*, Howard Chen*, Ryan Liu*, Manoel Horta Ribeiro, Robb Willer, Graham Neubig, and Thomas L. Griffiths
  Preprint 2025 [arXiv]


• LLM Social Simulations Are a Promising Research Method
  Jacy Reese Anthis, Ryan Liu, Sean M. Richardson, Austin C. Kozlowski, Bernard Koch, Erik Brynjolfsson, James Evans, and Michael Bernstein
  ICML 2025 Position Paper Track [arXiv]


• Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse
  Ryan Liu*, Jiayi Geng*, Addison J. Wu, Ilia Sucholutsky, Tania Lombrozo, and Thomas L. Griffiths
  ICML 2025 [arXiv]


• On Benchmarking Human-like Intelligence in Machines
  Lance Ying, Katherine M. Collins, Lionel Wong, Ilia Sucholutsky, Ryan Liu, Adrian Weller, Tianmin Shu, Thomas L. Griffiths, and Joshua B. Tenenbaum
  Preprint 2025 [arXiv]


• Large Language Models Assume People are More Rational than We Really Are
  Ryan Liu*, Jiayi Geng*, Joshua C. Peterson, Ilia Sucholutsky, and Thomas L. Griffiths
  ICLR 2025 [arXiv]


• Improving Interpersonal Communication by Simulating Audiences with Language Models
  Ryan Liu, Howard Yen, Raja Marjieh, Thomas L. Griffiths, and Ranjay Krishna
  CogSci 2025 [arXiv]


• LLMs as Workers in Human-Computational Algorithms? Replicating Crowdsourcing Pipelines with LLMs
  Tongshuang Wu, Haiyi Zhu, Maya Albayrak, Alexis Axon, Amanda Bertsch, Wenxing Deng, Ziqi Ding, Bill Guo, Sireesh Gururaja, Tzu-Sheng Kuo, Jenny T Liang, Ryan Liu, Ihita Mandal, Jeremiah Milbauer, Xiaolin Ni, Namrata Padmanabhan, Subhashini Ramkumar, Alexis Sudjianto, Jordan Taylor, Ying-Jui Tseng, Patricia Vaidos, Zhijin Wu, Wei Wu, and Chenyang Yang
  CHI 2025 Extended Abstract [arXiv]


• How do Large Language Models Navigate Conflicts between Honesty and Helpfulness?
  Ryan Liu*, Theodore R. Sumers*, Ishita Dasgupta, and Thomas L. Griffiths
  ICML 2024 Oral [arXiv]


• ReviewerGPT? An Exploratory Study on Using Large Language Models for Paper Reviewing
  Ryan Liu and Nihar B. Shah
  Oral, AAAI SDU Workshop 2024 [arXiv]


• API-Assisted Code Generation for Question Answering on Varied Table Structures
  Yihan Cao*, Shuyi Chen*, Ryan Liu*, Zhiruo Wang, and Daniel Fried
  EMNLP 2023 [arXiv]


• Testing for Reviewer Anchoring in Peer Review: A Randomized Controlled Trial
  Ryan Liu, Steven Jecmen, Fei Fang, Vincent Conitzer, and Nihar B. Shah
  PLoS ONE [arXiv]


• Cite-seeing and Reviewing: A Study on Citation Bias in Peer Review
  Ivan Stelmakh, Charvi Rastogi, Ryan Liu, Shuchi Chawla, Federico Echenique, and Nihar B. Shah
  PLoS ONE [arXiv]


• Near-Optimal Reviewer Splitting in Two-Phase Paper Reviewing & Conference Experiment Design
  Steven Jecmen, Hanrui Zhang, Ryan Liu, Fei Fang, Vincent Conitzer, and Nihar B. Shah
  AAAI HCOMP 2022, Best Paper Honorable Mention [arXiv]


• Mitigating Manipulation in Peer Review via Randomized Reviewer Assignments
  Steven Jecmen, Hanrui Zhang, Ryan Liu, Nihar B. Shah, Vincent Conitzer, and Fei Fang
  NeurIPS 2020 [arXiv]

Presentations

• Talk @ UCLA NLP Seminar
  The Psychology of AI Agents
• Talk @ Stanford Politics and Social Change Lab
  Predicting and Simulating New People using Existing Agents
• Talk @ The Stanford NLP Group Seminar
  Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse
• Talk @ Stanford HCI Group
  Predicting and Simulating New People using Existing Agents
• Talk @ UT-Austin Natural Language Learning Reading Group
  Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse
• Talk @ Google DeepMind
  Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse
• Oral @ International Conference on Machine Learning 2024
  How do Large Language Models Navigate Conflicts between Honesty and Helpfulness?
• Moderator @ London Machine Learning Meetup
  Joon Sung Park | Generative Agents: Interactive Simulacra of Human Behavior [recording]
• Podcast @ Data Skeptic
  Automated Peer Review [link]
• Visit @ Allen Institute for AI, Semantic Scholar Team
  
• Talk @ Carnegie Mellon University Meeting of the Minds 2022
  Identifying Human Biases in Peer Review via Real-Subject Experiments
• Poster @ Carnegie Mellon University Meeting of the Minds 2021
  Improving Algorithmic Tools for Conference Peer Review Research
• Poster @ Carnegie Mellon University Fall Undergraduate Research Showcase 2020
  Creating Robustness within Conference Peer Review
• Poster @ Carnegie Mellon University Meeting of the Minds 2020
  Assignment Algorithms to Prevent Quid-Pro-Quo in Conference Peer Review

Experience

• Assistant in Instruction @ Princeton | Advanced Topics in Computer Science: Machine Behavior
  
• Assistant in Instruction @ Princeton | Ethics of Computing
  
• AI/ML SWE Internship @ Meta
  
• Teaching Assistant @ CMU | 15-112 Fundamentals of Programming
  
• Research Assistant @ CMU School of Computer Science
  

Academic Honors

• Reviewer, NeurIPS 2026 Position Paper Track
  
• Reviewer, COLM 2026
  
• Reviewer, ICML 2026
  
• Reviewer, ICLR 2026
  
• Reviewer, NeurIPS 2025 Workshop on CogInterp: Interpreting Cognition in Deep Learning Models
  
• Reviewer, NeurIPS 2025
  
• Reviewer, ICLR 2025 Workshop on Bidirectional Human-AI Alignment
  
• Reviewer, ICLR 2025
  
• Reviewer, NeurIPS 2024 Workshop on Behavioral ML
  
• Reviewer, NeurIPS 2024
  
• Student Organizer, Decentralized Social Media Workshop @Princeton
  
• NSF Research Experience for Undergraduates Grant (CMU)
  
• Bachelor of Science, CMU School of Computer Science, College & University Honors
  
• Fifth-Year Master's, CMU School of Computer Science, Thesis: Testing for Reviewer Anchoring in the Conference Rebuttal Process [link]